I'm trying to make my own Move to Resource refactoring (like the one found in ReSharper). In my CodeAction.GetChangedDocumentAsync method I'm doing these 3 steps:
Add my resource to the Resources.resx file with XmlDocument
Use DTE to run the custom tool to update the Resources.Designer.cs file
Replace the literal string by the qualified identifier of the new resource with SyntaxNode.ReplaceNode
Steps 1 & 2 are OK, but 3 does not work. If I remove the step 2, 3 is working.
I don't know if it is because I'm mixing Roslyn & DTE or if it's because step 2 generate new code in the solution and my cached context becomes invalid.
// Add the resource to the Resources.resx file
var xmlDoc = new XmlDocument();
xmlDoc.Load(resxPath);
XmlNode node = xmlDoc.SelectSingleNode($"//data[#name='{resourceIndentifierName}']");
if (node != null) return;
XmlElement dataElement = xmlDoc.CreateElement("data");
XmlAttribute nameAtt = xmlDoc.CreateAttribute("name");
nameAtt.Value = resourceIndentifierName;
dataElement.Attributes.Append(nameAtt);
XmlAttribute spaceAtt = xmlDoc.CreateAttribute("space", "xml");
spaceAtt.Value = "preserve";
dataElement.Attributes.Append(spaceAtt);
XmlElement valueElement = xmlDoc.CreateElement("value");
valueElement.InnerText = value;
dataElement.AppendChild(valueElement);
XmlNode rootNode = xmlDoc.SelectSingleNode("//root");
Debug.Assert(rootNode != null, "rootNode != null");
rootNode.AppendChild(dataElement);
xmlDoc.Save(resxPath);
// Update the Resources.Designer.cs file
var dte = (DTE2)Package.GetGlobalService(typeof(SDTE));
ProjectItem item = dte.Solution.FindProjectItem(resxPath);
((VSProjectItem) item.Object)?.RunCustomTool();
// Replace the node
SyntaxNode oldRoot = await context.Document.GetSyntaxRootAsync(cancellationToken)
.ConfigureAwait(false);
SyntaxNode newRoot = oldRoot.ReplaceNode(oldNode, newNode);
return context.Document.WithSyntaxRoot(newRoot);
it's because step 2 generate new code in the solution and my cached context becomes invalid
This is what is happening. When you call RunCustomTool the visual studio file watcher api tells roslyn that the files have been updated and roslyn generates a new set of solution snapshots. When you attempt to apply your codefix, roslyn looks at the solution snapshot that your codefix came from, sees that it does not match the current snapshot, and fails to apply it.
Related
I have an XML file that collects information with Button_Click, so it starts off empty.
XML Sample
<marina>
<dockone>
</dockone>
<docktwo>
</docktwo>
</marina>
When I submit information from a textbox, a new XmlNode is created called slipone, and another XmlNode called reg is nested within that.
XML Sample 2
<marina>
<dockone>
<slipone>
<reg>12345</reg>
<slipone>
</dockone>
<docktwo>
</docktwo>
</marina>
I have attempted to create an if/else statement that will add a new XmlNode called sliptwo, with reg still nested within it, if slipone already has text, like so:
<marina>
<dockone>
<slipone>
<reg>12345</reg>
<slipone>
<sliptwo>
<reg>67890</reg>
<sliptwo>
</dockone>
<docktwo>
</docktwo>
</marina>
However the closest I have gotten is another XMlnode is still created, however it labels itself as slipone, and I am not sure what I am doing wrong:
<marina>
<dockone>
<slipone>
<reg>12345</reg>
<slipone>
<slipone>
<reg>67890</reg>
<slipone>
</dockone>
<docktwo>
</docktwo>
</marina>
This is an example of what I have been playing around with. Ignore the operators as I have resorted to trial and error but still have gotten nowhere. Please help!
C# Example
XmlDocument XmlDocObj1 = new XmlDocument();
XmlDocObj1.Load(Server.MapPath("~/App_Data/SlipData.xml"));
XmlNode rootnode1 = XmlDocObj1.SelectSingleNode("marina/dockone");
XmlNode dockone = rootnode1.AppendChild(XmlDocObj1.CreateNode(XmlNodeType.Element, "slipone", ""));
XmlNode docktwo = rootnode1.AppendChild(XmlDocObj1.CreateNode(XmlNodeType.Element, "sliptwo", ""));
XmlNode dockthree = rootnode1.AppendChild(XmlDocObj1.CreateNode(XmlNodeType.Element, "slipthree", ""));
if (regfinal.Text != dockone.InnerText)
{
dockone.AppendChild(XmlDocObj1.CreateNode(XmlNodeType.Element, "Reg", "")).InnerText = regfinal.Text;
XmlDocObj1.Save(Server.MapPath("/App_Data/SlipData.xml"));
}
else if (regfinal.Text == dockone.InnerText)
{
docktwo.AppendChild(XmlDocObj1.CreateNode(XmlNodeType.Element, "Reg", "")).InnerText = regfinal.Text;
XmlDocObj1.Save(Server.MapPath("/App_Data/SlipData.xml"));
}
Your logic isn't going to do what I think you are saying since the only time (regfinal.Text != dockone.InnerText) will evaluate to false is when you enter nothing in your text control.
I believe you might mean to say if dockone exists then create another node called docktwo. This will require you to change your logic.
Some very simple code to get you a bit farther down the path. Not intended to be perfect or solve all problems...
private void button1_Click(object sender, EventArgs e)
{
XmlDocument XmlDocObj1 = new XmlDocument();
XmlDocObj1.Load(AppDomain.CurrentDomain.BaseDirectory.ToString()+"test.xml");
XmlNode rootnode1 = XmlDocObj1.SelectSingleNode("marina/dockone");
XmlNode dockone = rootnode1.AppendChild(XmlDocObj1.CreateNode(XmlNodeType.Element, "slipone", ""));
XmlNode docktwo = rootnode1.AppendChild(XmlDocObj1.CreateNode(XmlNodeType.Element, "sliptwo", ""));
XmlNode dockthree = rootnode1.AppendChild(XmlDocObj1.CreateNode(XmlNodeType.Element, "slipthree", ""));
//jsh: old logic
//if (textBox1.Text != dockone.InnerText)
//new logic to test whether we have already created the dockone node which should only occur once
//you already have the logic for selecting the dockone node above...now just test if you already have it.
//NOTE: you may actually want a switch statement given that you avhe dockone, docktwo, and dockthree or at least another
// if statement to see if docktwo has been created and thus creaste dockthree.
if (rootnode1 == null )
{
dockone.AppendChild(XmlDocObj1.CreateNode(XmlNodeType.Element, "Reg", "")).InnerText = textBox1.Text;
XmlDocObj1.Save(AppDomain.CurrentDomain.BaseDirectory.ToString() + "test.xml");
}
//else if (textBox1.Text == dockone.InnerText) jsh: old logic
else
{
docktwo.AppendChild(XmlDocObj1.CreateNode(XmlNodeType.Element, "Reg", "")).InnerText = textBox1.Text;
XmlDocObj1.Save(AppDomain.CurrentDomain.BaseDirectory.ToString() + "test.xml");
}
}
I'm trying to find out how to read/write to the extended file properties in C#
e.g. Comment, Bit Rate, Date Accessed, Category etc that you can see in Windows explorer.
Any ideas how to do this?
EDIT: I'll mainly be reading/writing to video files (AVI/DIVX/...)
For those of not crazy about VB, here it is in c#:
Note, you have to add a reference to Microsoft Shell Controls and Automation from the COM tab of the References dialog.
public static void Main(string[] args)
{
List<string> arrHeaders = new List<string>();
Shell32.Shell shell = new Shell32.Shell();
Shell32.Folder objFolder;
objFolder = shell.NameSpace(#"C:\temp\testprop");
for( int i = 0; i < short.MaxValue; i++ )
{
string header = objFolder.GetDetailsOf(null, i);
if (String.IsNullOrEmpty(header))
break;
arrHeaders.Add(header);
}
foreach(Shell32.FolderItem2 item in objFolder.Items())
{
for (int i = 0; i < arrHeaders.Count; i++)
{
Console.WriteLine(
$"{i}\t{arrHeaders[i]}: {objFolder.GetDetailsOf(item, i)}");
}
}
}
Solution 2016
Add following NuGet packages to your project:
Microsoft.WindowsAPICodePack-Shell by Microsoft
Microsoft.WindowsAPICodePack-Core by Microsoft
Read and Write Properties
using Microsoft.WindowsAPICodePack.Shell;
using Microsoft.WindowsAPICodePack.Shell.PropertySystem;
string filePath = #"C:\temp\example.docx";
var file = ShellFile.FromFilePath(filePath);
// Read and Write:
string[] oldAuthors = file.Properties.System.Author.Value;
string oldTitle = file.Properties.System.Title.Value;
file.Properties.System.Author.Value = new string[] { "Author #1", "Author #2" };
file.Properties.System.Title.Value = "Example Title";
// Alternate way to Write:
ShellPropertyWriter propertyWriter = file.Properties.GetPropertyWriter();
propertyWriter.WriteProperty(SystemProperties.System.Author, new string[] { "Author" });
propertyWriter.Close();
Important:
The file must be a valid one, created by the specific assigned software. Every file type has specific extended file properties and not all of them are writable.
If you right-click a file on desktop and cannot edit a property, you wont be able to edit it in code too.
Example:
Create txt file on desktop, rename its extension to docx. You can't
edit its Author or Title property.
Open it with Word, edit and save
it. Now you can.
So just make sure to use some try catch
Further Topic:
Microsoft Docs: Implementing Property Handlers
There's a CodeProject article for an ID3 reader. And a thread at kixtart.org that has more information for other properties. Basically, you need to call the GetDetailsOf() method on the folder shell object for shell32.dll.
This sample in VB.NET reads all extended properties:
Sub Main()
Dim arrHeaders(35)
Dim shell As New Shell32.Shell
Dim objFolder As Shell32.Folder
objFolder = shell.NameSpace("C:\tmp")
For i = 0 To 34
arrHeaders(i) = objFolder.GetDetailsOf(objFolder.Items, i)
Next
For Each strFileName In objfolder.Items
For i = 0 To 34
Console.WriteLine(i & vbTab & arrHeaders(i) & ": " & objfolder.GetDetailsOf(strFileName, i))
Next
Next
End Sub
You have to add a reference to Microsoft Shell Controls and Automation from the COM tab of the References dialog.
Thank you guys for this thread! It helped me when I wanted to figure out an exe's file version. However, I needed to figure out the last bit myself of what is called Extended Properties.
If you open properties of an exe (or dll) file in Windows Explorer, you get a Version tab, and a view of Extended Properties of that file. I wanted to access one of those values.
The solution to this is the property indexer FolderItem.ExtendedProperty and if you drop all spaces in the property's name, you'll get the value. E.g. File Version goes FileVersion, and there you have it.
Hope this helps anyone else, just thought I'd add this info to this thread. Cheers!
GetDetailsOf() Method - Retrieves details about an item in a folder. For example, its size, type, or the time of its last modification. File Properties may vary based on the Windows-OS version.
List<string> arrHeaders = new List<string>();
Shell shell = new ShellClass();
Folder rFolder = shell.NameSpace(_rootPath);
FolderItem rFiles = rFolder.ParseName(filename);
for (int i = 0; i < short.MaxValue; i++)
{
string value = rFolder.GetDetailsOf(rFiles, i).Trim();
arrHeaders.Add(value);
}
Jerker's answer is little simpler. Here's sample code which works from MS:
var folder = new Shell().NameSpace(folderPath);
foreach (FolderItem2 item in folder.Items())
{
var company = item.ExtendedProperty("Company");
var author = item.ExtendedProperty("Author");
// Etc.
}
For those who can't reference shell32 statically, you can invoke it dynamically like this:
var shellAppType = Type.GetTypeFromProgID("Shell.Application");
dynamic shellApp = Activator.CreateInstance(shellAppType);
var folder = shellApp.NameSpace(folderPath);
foreach (var item in folder.Items())
{
var company = item.ExtendedProperty("Company");
var author = item.ExtendedProperty("Author");
// Etc.
}
After looking at a number of solutions on this thread and elsewhere
the following code was put together. This is only to read a property.
I could not get the
Shell32.FolderItem2.ExtendedProperty function to work, it is supposed
to take a string value and return the correct value and type for that
property... this was always null for me and developer reference resources were very thin.
The WindowsApiCodePack seems
to have been abandoned by Microsoft which brings us the code below.
Use:
string propertyValue = GetExtendedFileProperty("c:\\temp\\FileNameYouWant.ext","PropertyYouWant");
Will return you the value of the extended property you want as a
string for the given file and property name.
Only loops until it found the specified property - not until
all properties are discovered like some sample code
Will work on Windows versions like Windows server 2008 where you will get the error "Unable to cast COM object of type 'System.__ComObject' to interface type 'Shell32.Shell'" if just trying to create the Shell32 Object normally.
public static string GetExtendedFileProperty(string filePath, string propertyName)
{
string value = string.Empty;
string baseFolder = Path.GetDirectoryName(filePath);
string fileName = Path.GetFileName(filePath);
//Method to load and execute the Shell object for Windows server 8 environment otherwise you get "Unable to cast COM object of type 'System.__ComObject' to interface type 'Shell32.Shell'"
Type shellAppType = Type.GetTypeFromProgID("Shell.Application");
Object shell = Activator.CreateInstance(shellAppType);
Shell32.Folder shellFolder = (Shell32.Folder)shellAppType.InvokeMember("NameSpace", System.Reflection.BindingFlags.InvokeMethod, null, shell, new object[] { baseFolder });
//Parsename will find the specific file I'm looking for in the Shell32.Folder object
Shell32.FolderItem folderitem = shellFolder.ParseName(fileName);
if (folderitem != null)
{
for (int i = 0; i < short.MaxValue; i++)
{
//Get the property name for property index i
string property = shellFolder.GetDetailsOf(null, i);
//Will be empty when all possible properties has been looped through, break out of loop
if (String.IsNullOrEmpty(property)) break;
//Skip to next property if this is not the specified property
if (property != propertyName) continue;
//Read value of property
value = shellFolder.GetDetailsOf(folderitem, i);
}
}
//returns string.Empty if no value was found for the specified property
return value;
}
Here is a solution for reading - not writing - the extended properties based on what I found on this page and at help with shell32 objects.
To be clear this is a hack. It looks like this code will still run on Windows 10 but will hit on some empty properties. Previous version of Windows should use:
var i = 0;
while (true)
{
...
if (String.IsNullOrEmpty(header)) break;
...
i++;
On Windows 10 we assume that there are about 320 properties to read and simply skip the empty entries:
private Dictionary<string, string> GetExtendedProperties(string filePath)
{
var directory = Path.GetDirectoryName(filePath);
var shell = new Shell32.Shell();
var shellFolder = shell.NameSpace(directory);
var fileName = Path.GetFileName(filePath);
var folderitem = shellFolder.ParseName(fileName);
var dictionary = new Dictionary<string, string>();
var i = -1;
while (++i < 320)
{
var header = shellFolder.GetDetailsOf(null, i);
if (String.IsNullOrEmpty(header)) continue;
var value = shellFolder.GetDetailsOf(folderitem, i);
if (!dictionary.ContainsKey(header)) dictionary.Add(header, value);
Console.WriteLine(header +": " + value);
}
Marshal.ReleaseComObject(shell);
Marshal.ReleaseComObject(shellFolder);
return dictionary;
}
As mentioned you need to reference the Com assembly Interop.Shell32.
If you get an STA related exception, you will find the solution here:
Exception when using Shell32 to get File extended properties
I have no idea what those properties names would be like on a foreign system and couldn't find information about which localizable constants to use in order to access the dictionary. I also found that not all the properties from the Properties dialog were present in the dictionary returned.
BTW this is terribly slow and - at least on Windows 10 - parsing dates in the string retrieved would be a challenge so using this seems to be a bad idea to start with.
On Windows 10 you should definitely use the Windows.Storage library which contains the SystemPhotoProperties, SystemMusicProperties etc.
https://learn.microsoft.com/en-us/windows/uwp/files/quickstart-getting-file-properties
And finally, I posted a much better solution that uses WindowsAPICodePack there
I'm not sure what types of files you are trying to write the properties for but taglib-sharp is an excellent open source tagging library that wraps up all this functionality nicely. It has a lot of built in support for most of the popular media file types but also allows you to do more advanced tagging with pretty much any file.
EDIT: I've updated the link to taglib sharp. The old link no longer worked.
EDIT: Updated the link once again per kzu's comment.
I'm trying to get the current iteration path for the teams TFS project. The way I'm trying to do this is by using the blog from http://blog.johnsworkshop.net/tfs11-api-reading-the-team-configuration-iterations-and-areas/ . I start by getting the team configurations from the following code:
TfsTeamProjectCollection tpc = TFSConncetion(#"http://tfs/url");
var configSvc = tpc.GetService<TeamSettingsConfigurationService>();
var configs = configSvc.GetTeamConfigurationsForUser(projectUri);
The problem with this is that my configs is always null, even though I'm a member of the team. I'm positive my projects URI is correct as well. After this I would get the team settings and use that to display the current iteration path.
TeamSettings ts = config.TeamSettings;
Console.WriteLine(ts.CurrentIterationPath);
Even if this didn't work I can still query the iteration dates from the team setting to get the one iteration that has a start date before today and finish date after today. The main problem is that I can't get my TeamSettingsConfigurationService to return anything but null when I try to get the team configurations with my projects URI.
There must be something wrong with your server connection or the project uri you're passing as the other code looks okay.
Maybe try something like this:
TfsTeamProjectCollection tpc = new TfsTeamProjectCollection(new Uri("http://server:8080/tfs/collection"),
new System.Net.NetworkCredential(tfsUserName, tfsPassword));
tpc.EnsureAuthenticated();
Connect to Team Foundation Server from a Console Application
There is a good sample here which you can download (WPF client) and it will allow you to select a server connection, Team Project and Team:
TFS API Part 46 (VS11) – Team Settings
You can step through it and check the values you're passing into your code.
The sample gets the team configuration information is the same way you have in your code.
TeamSettingsConfigurationService teamConfig = tfs.GetService<TeamSettingsConfigurationService>();
var configs = teamConfig.GetTeamConfigurationsForUser(new[] { projectInfo.Uri });
Once you have the collection of TeamConfiguration items then you need TeamSettings.CurrentIterationPath
I actually got the answer myself without using TeamSettingsConfigurationService at all. Here's how I did it:
private static XmlNode currentIterationNode;
TfsTeamProjectCollection tpc = TFSConncetion(#"http://tfs/url");
ICommonStructureService4 css = tpc.GetService<ICommonStructureService4>();;
WorkItemStore workItemStore = new WorkItemStore(tpc);
foreach (Project teamProject in workItemStore.Projects)
{
if (teamProject.Name.Equals("TeamProjectNameGoesHere"))
{
NodeInfo[] structures = css.ListStructures(teamProject.Uri.ToString());
NodeInfo iterations = structures.FirstOrDefault(n => n.StructureType.Equals("ProjectLifecycle"));
if (iterations != null)
{
XmlElement iterationsTree = css.GetNodesXml(new[] { iterations.Uri }, true);
XmlNodeList nodeList = iterationsTree.ChildNodes;
currentIterationNode = FindCurrentIteration(nodeList);
String currentIterationPath = currentIterationNode.Attributes["Path"].Value;
}
}
}
Where currentIterationPath is the current iteration path from TFS. The key to doing this was to get the NodeInfo[] array of structures and the NodeInfo iterations from these two lines of code I got from chamindacNavantis https://social.msdn.microsoft.com/Forums/vstudio/en-US/4b785ae7-66c0-47ee-a6d2-c0ad8a3bd420/tfs-get-iteration-dates-metadata?forum=tfsgeneral:
NodeInfo[] structures = css.ListStructures(teamProject.Uri.ToString());
NodeInfo iterations = structures.FirstOrDefault(n => n.StructureType.Equals("ProjectLifecycle"));
After that I created an xml with nodes of every iteration inside the team project. These nodes also have the start date and end dates of each iteration. So I checked each node for a start date before DateTime.Now and finish date after DateTime.Now, which is all FindCurrentIteration(nodeList) does.
And that will give you the current iteration node.
The simplest way I've found to do it was by using ICommonStructureService4 and TeamSettingsConfigurationService methods:
static TfsTeamProjectCollection _tfs = TfsTeamProjectCollectionFactory
.GetTeamProjectCollection("<tfsUri>")
(...)
static string GetCurrentIterationPath()
{
var css = _tfs.GetService<ICommonStructureService4>();
var teamProjectName = "<teamProjectName>";
var project = css.GetProjectFromName(teamProjectName);
var teamName = "<teamName>";
var teamSettingsStore = _tfs.GetService<TeamSettingsConfigurationService>();
var settings = teamSettingsStore
.GetTeamConfigurationsForUser(new[] { project.Uri })
.Where(c => c.TeamName == teamName)
.FirstOrDefault();
if (settings == null)
{
var currentUser = System.Threading.Thread.CurrentPrincipal.Identity.Name;
throw new InvalidOperationException(
$"User '{currentUser}' doesn't have access to '{teamName}' team project.");
}
return settings.TeamSettings.CurrentIterationPath;
}
I'm trying to update an existing XML file, but always when I update it adding new tags the xmlns="" attribute mysteriously appears in all tags and I didn't find a way to remove it.
private static void EditarXML(string path, List<SiteUrl> listaUrls, bool indice, string loc)
{
XmlDocument documentoXML = new XmlDocument();
documentoXML.Load(path);
XmlNode sitemap = documentoXML.CreateElement("sitemap");
XmlNode xloc = documentoXML.CreateElement("loc");
xloc.InnerText = loc;
sitemap.AppendChild(xloc);
XmlNode lastmod = documentoXML.CreateElement("lastmod");
lastmod.InnerText = DateTime.Now.ToShortDateString();
sitemap.AppendChild(lastmod);
documentoXML.DocumentElement.AppendChild(sitemap);
}
Any help or ideas would be appreciated.
This will happen with the parent node you are appending to has a namespace, but you don't specify it in the CreateElement() call.
To handle this, you can get the namespace from the DocumentElement, like this (my sample just creates the document in memory, but the principle is the same), and pass it to CreateElement().
if (x.DocumentElement != null) {
var xmlns = (x.DocumentElement.NamespaceURI);
var sitemap = x.CreateElement("sitemap", xmlns);
var xloc = x.CreateElement("loc", xmlns);
xloc.InnerText = "Hello";
sitemap.AppendChild(xloc);
var lastmod = x.CreateElement("lastmod", xmlns);
lastmod.InnerText = DateTime.Now.ToShortDateString();
sitemap.AppendChild(lastmod);
x.DocumentElement.AppendChild(sitemap);
}
Console.WriteLine(x.InnerXml);
Output
<test xmlns="jdphenix"><sitemap><loc>Hello</loc><lastmod>4/20/2015</lastmod></sitemap></test>
Note that if I did not pass the parent namespace to each CreateElement() call, the children of that call would have the blank xmlns.
// incorrect - appends xmlns=""
if (x.DocumentElement != null) {
var sitemap = x.CreateElement("sitemap");
var xloc = x.CreateElement("loc");
xloc.InnerText = "Hello";
sitemap.AppendChild(xloc);
var lastmod = x.CreateElement("lastmod");
lastmod.InnerText = DateTime.Now.ToShortDateString();
sitemap.AppendChild(lastmod);
x.DocumentElement.AppendChild(sitemap);
}
Console.WriteLine(x.InnerXml);
Output
<test xmlns="jdphenix"><sitemap xmlns=""><loc>Hello</loc><lastmod>4/20/2015</lastmod></sitemap></test>
Related reading: Why does .NET XML append an xlmns attribute to XmlElements I add to a document? Can I stop it?
How to prevent blank xmlns attributes in output from .NET's XmlDocument?
I'm currently trying to create a custom tree and I'm running into trouble when trying to render a nodes children. After browsing various articles/posts I'm at this point:
public override void Render(ref XmlTree tree)
{
List<Node> articles = NodeUtil.GetAllNodesOfDocumentType(-1, "Promoter");
Node article = articles.Where(p => p.CreatorID == UmbracoEnsuredPage.CurrentUser.Id).FirstOrDefault();
if(promo != null)
{
var dNode = XmlTreeNode.Create(this);
dNode.NodeID = article.Id.ToString();
dNode.Text = article.Name;
dNode.Icon = "doc.gif";
dNode.Action = "javascript:openArticle(" + article.Id + ")";
dNode.Source = article.Children.Count > 0 ? this.GetTreeServiceUrl("" + article.Id) : "";
tree.Add(dNode);
}
}
The code above gets the article belonging to the current user (for the sake of testing, each user only has one article at the moment). I then attempt to print out the children of this article but instead of getting the desired output, I get the follwowing:
Article Name
- Article Name
- Article Name
- Article Name
Each time I expand a node, it just seems to render the same node, and goes on and on.
I've seen other ways of using the treeservice, like:
TreeService treeService = new TreeService(...);
node.Source = treeService.GetServiceUrl();
But I get an error saying there is no GetServiceUrl method that takes 0 arguments. I assume the method above was for earlier versions?
It took me a while to work this out. Here is the solution, hope it would help someone.
const string PARENT_ID = "10"; // The ID of the node that has child nodes
public override void Render(ref XmlTree tree)
{
if (this.NodeKey == PARENT_ID) // Rendering the child nodes of the parent folder
{
// Render a child node
XmlTreeNode node = XmlTreeNode.Create(this);
node.NodeID = "11";
node.Text = "child";
node.Icon = "doc.gif";
node.Action = ...
tree.Add(node);
}
else // Default (Rendering the root)
{
// Render the parent folder
XmlTreeNode node = XmlTreeNode.Create(this);
node.NodeID = PARENT_ID;
node.Source = this.GetTreeServiceUrl(node.NodeID);
node.Text = "parent";
node.Icon = "folder.gif";
tree.Add(node);
}
}
The output suggests that the node tree you're building is nesting each child node - this is because the nodeId is being reset to -1 with each pass.
This post on our.umbraco.org describes the same problem, and suggests that you use NodeKey instead of ID to move between nodes.
**
Not necessarily helpful but I would use the uQuery language extensions that comes with the ucomponents package (and who installs Umbraco without ucomponents?), to simplify the method calls:
For example:
List<Node> articles = uQuery.getNodesByType("Promoter");
foreach(Node article in articles)
{
List<Node> children = article.GetDescendantNodes();
... build tree
}